CVLGOct 5, 2018

Hierarchical Recurrent Filtering for Fully Convolutional DenseNets

arXiv:1810.02766v21 citations
Originality Incremental advance
AI Analysis

This work addresses perception failures in challenging situations for learning agents, but it is incremental as it builds on existing single-frame models.

The authors tackled the problem of robust environment representation for learning agents by extending a single-frame segmentation model to handle multiple frames with a parameter-efficient temporal filtering concept, resulting in a hierarchical recurrent architecture that decouples temporal dependencies from scene representation and demonstrated improved ability to cope with data perturbations on a synthetic dataset.

Generating a robust representation of the environment is a crucial ability of learning agents. Deep learning based methods have greatly improved perception systems but still fail in challenging situations. These failures are often not solvable on the basis of a single image. In this work, we present a parameter-efficient temporal filtering concept which extends an existing single-frame segmentation model to work with multiple frames. The resulting recurrent architecture temporally filters representations on all abstraction levels in a hierarchical manner, while decoupling temporal dependencies from scene representation. Using a synthetic dataset, we show the ability of our model to cope with data perturbations and highlight the importance of recurrent and hierarchical filtering.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes